Reduction at the Streaming Mercury Electrode. 11. Current-Voltage

The data in Table I1 for cyclohexene at 0 and 25' indicate that ... steady-state current at the streaming electrode ... constant voltage, ideal stream...
1 downloads 3 Views 2MB Size
5542

J. R. WEAVERAND R. W. PARRY

inum-black electrode, the hydrogenation could be diffusion-controlled rather than kinetically controlled. This possibility was checked by determining the rate constant for cyclohexene a t 0' as well as a t 25'. Diffusion coefficients normally change 1% per degree centigrade, while reaction rates increase much more rapidly with temperature. The data in Table I1 for cyclohexene at 0 and 25' indicate that the change in rate is too large to be solely diffusion controlled. Furthermore, a change in hydrogen concentration on the catalytic surface causes a like change in the rate of reaction. If the reaction were diffusion controlled, no change in rate would be observed since the reaction must take place a t the electrode. An attempt was made t o study benzene, but its reaction rate was too slow t o be observable above the blank generation rate. Studies of styrene were also unsatisfactory because of low solubility and polymerization. Summary.-Relative rates of hydrogenation can be determined rapidly and conveniently with this method. It should be emphasized that the concentrations of hydrogen are much lower than with conventional methods. Rate constants which are determined coulometrically may be of more

[CONTRIBUTION FROM THE

Vol.

'is

interest than those evaluated by allowing the reaction to proceed to infinite time. By determining the rate of reaction a t constant concentrations of reactants the problem of change of rate law with changes of concentration is avoided. The conditions are also more closely related to industrial continuous stream processes. The limitations for applying the method are : 1, sufficient reactivity for the olefins; 2, sufficient solubility of the olefin; 3, sufficiently low vapor pressure for the olefin to prevent loss of sample. Highly active olefins may be studied by using lower concentrations of hydrogen, and comparing the generation rate with that of a known olefin a t the lower concentration. Olefins of low reactivity cannot be studied because the concentrations of hydrogen are too low t o produce sufficiently rapid reaction. The method has applications other than in the determination of the relative reactivities of olefins. By studying the same olefin, the relative effective area of different samples of a given type of catalyst and the relative reactivities of different catalysts can be determined. I n a similar way, the effects of differentsolvents might be studied. LOSANGELES24, CALIFORNIA

DEPARTMENT O F CHEMISTRY

O F THE UNIVERSITY O F MICHIGAN]

Reduction at the Streaming Mercury Electrode. 11. Current-Voltage Curves' BY J. R. WEAVERAND R. W. PARRY RECEIVED APRIL3, 1956

A method has been developed for treating current-voltage curves obtained from steady-state measurements a t the streaming mercury electrode. Applications to experimental curves using thallium, cadmium and lead ions for reversible reduction and zinc ion for irreversible reduction are shown. Rate constants for the reduction of aqueous solutions of zinc ion in several concentrations of potassium chloride are reported.

I n a previous paper12 studies of the limiting steady-state current a t the streaming electrode were reported with emphasis on factors causing deviations from the Rius e q ~ a t i o n . ~In this paper current-voltage curves obtained from steadystate measurements will be considered. Because of the short time of contact between mercury and solution, a reduction process showing a tendency toward irreversibility should give a current-voltage curve with a greater deviation from the reversible form than it gives a t the dropping electrode. The streaming electrode should thus prove particularly useful in investigating kinetic features of electrode processes, providing the currentvoltage measurements can be properly interpreted. A treatment for irreversible reduction a t the streaming electrode has been given by Koryta.* Implicit in his derivation are the assumptions, first, that the physical characteristics of the stream (1) Abstracted in part from a dissertation submitted by J . R. Weaver t o the Horace H. Rackham School of Graduate Studies of the University of Michigan in partial fulfillment of the requirements for the degree of Doctor of Philosophy. (2) J. R . Weaver and R . W. Parry, THIS JOURNAL, 76, 6258 (1954). (3) A. Rius, J. Llopis and S. Polo, A n d e s real soc. Espan. 3 s . y q?ri?n., 46B, 1039 (1949). (4) J . Koryta, Coli. Czech. Chem. Comm., 19, 483 (1964).

are those of the hypothetical "ideal streaming electrode," i.e., constant radius, constant surface velocity, and negligible effects from electrode curvature and velocity gradient; second, that the reduction potential does not vary with position on the stream; and third, that the effective length of the stream remains constant. Under these conditions an element of surface neither expands nor contracts as i t moves forward on the electrode, and reduction occurs exactly as a t a plane, stationary, constant-voltage electrode. For an element which has been in existence for t seconds, the current density, i, is given byjF7 i

=

nFCk exp

(g)

erfc

(k Kd')D

(1)

in which

and C is the bulk concentration of the reducible species, k and k b are the first-order] heterogeneous rate constants for the forward and reverse reactions, and D and D, are the diffusion coefficients (5) J . Koutecky and R . Brdicka, ibid., 12, 337 (1947). (6) P . Delahay, THISJOURNAL, 76, 1430 (1953). (7) D. M. H.Kern, ibid., 76, 2473 (1953).

REDUCTION AT

Nov. 5 , 1956

THE

STREAMING MERCURY ELECTRODE

of the reducible and reduced species. For an electrode of length L, surface velocity v, and radius 7 , the total current is

I

= 2 ~ r z ‘ r dt i =

4 ~ z F C r d x K{ 1 exp

where

4; 2Y

-11

(r2)erfc (Y)II

L

k

Y = zdD;

(3)

(4)

For convenience the bracketed expression will be designated asf(y)

4/71. - exp(y2) erfc(r)l f(r) = 1 - ---I1 2Y

(5)

and introducing the expression for the limiting current2 ID = 4nFCr.\/*DLv (6) equation 3 can be written as 1

= KAY)

(7)

This is the general equation for reduction a t the constant voltage, ideal streaming electrode. The distinguishing feature of reversible reduction is that, a t every point on the reduction wave, y is sufficiently large for f(y) to be unity within the experimental error of the measurements. Thus I-r e- , E

K

( 8) ID which for equilibrium conditions, is equivalent to RT I E = E l / %- (9) nF log, ID - I where El/, is the conventional half-wave potential. If K retains the equilibrium value for irreversible reduction as well, equation 7 may be written I - = f(Y) (10) I,,” where I r e v is the current that would be drawn by the same electrode process if i t were occurring reversibly, e.g., if accelerated by catalysis. It appears, however, as pointed out in the final section, that this condition does not hold for the example of irreversible reduction studied in this investigation. In any case, for “totally irreversible” reduction, where the reverse process has ceased, one can write I

=f(r)

A tabulation of the function f(y) is shown in Table V. The Non-ideal, Constant-Voltage Electrode.I n order to investigate the effects on currentvoltage curves which are attributable to deviations from “ideal” electrode behavior, the solution of the diffusion problem a t a real streaming electrode was carried through in a manner similar t o that given in the first paper of this series, but with the inclusion of a differential equation for diffusion of the reduced metal into the amalgam. For reversible reduction, the result obtained was

h =K ( l - (1 - K ) A + . . . . . I ID

(11)

5543

where A is given by

(12)

in which vl is the velocity gradient in the solution near the interface and y is the ratio of the viscosity of the solution to that of mercury. For the log plot one obtains K IW = log,--- A ... 1% 7 1 - K ID I,,, Three significant observations can be noted from these equations: (1) the factors which caused nearly all of the deviation of the limiting current from the Rius equation do not affect the shape of the reduction wave. This is shown by the fact that the principal term of equation 11 involves neither v nor r . An integral expression involving these quantities cancelled out upon division by ID indicating that the effect of the variations in stream radius and surface velocity are, proportionately, the same at all points on the wave.8 (2) I n the expression for A , the first term in the bracket represents the contribution of the curvature of the electrode and the second of the velocity gradient. These two terms are of the same order of magnitude and, since vl is negative, they tend to cancel each other. For the stream used in this investigation, the second term in equation 11 contributed, a t its maximum value, less than 1% to the current ratio. (3) The voltage-variable quantity K does not appear in the second term on the right-hand side of equation 13. This means that the correction term does not alter either the linearity or slope of the log plot. The only effect is a shift in the half-wave potential, which, for substances with diffusion coefficients of the order of cm.2/sec., is about 1 mv. in the negative direction. It was concluded that for reversible reduction the departure from ideal behavior does not seriously alter the reduction wave, and that, for a fixed stream length, i t is possible to define a hypothetical ideal electrode equivalent to the real electrode. The choice is not unique, since, as shown by equation 6 and the preceding discussion, all ideal electrodes with the same value for r v ’ i are equivalent, for reversible reduction. At the opposite extreme from reversible reduction is the state that will be designated as “complete rate-control.” It may be defined by the condition that y '

' '

1

' '

1

'

I

18 '

IL I', ,4' Ib ' 210 8 2 w 2 ! l 6 STREAM LENGTH FROM ORIFICE TO BASE OF TAILCONE.".

9

MERCURY

ELECTRODE

5545

-85

0

2

2

3

4

5

6

7

8

9

1

O

l

l

l

2

Distance from orifice, mm. Fig. 2.-Variation in electrode potential with distance from the orifice for a typical reduction situation. (Calculated by the method described in the section on variable ID= 2000 voltage electrode for E,n = -0.850 v. os. S.C.E., fi amp., .Ece,,= -0.906 v.; inert electrolyte, 0.1 M KC1.)

Fig. 1.-Cell conductance with 0.1 M KCI as determined by an alternating current method ( 0 )and an oscillographic method (0).

tion. Inserting also the expression for the limiting current from equation 6 gives

The corresponding expression for the concentration of the reduced species a t the interface is

These equations depend only on diffusion and are independent of the type of electrode process. For reversible reduction, the electrode voltage was obtained from the Nernst equation in the form

For irreversible reduction, a rate equation of the 0

with arbitrary values for CY and ko was used. Equations 18, 19 and either 20 or 21 were used to obtain the electrode potential from the arbitrary current density distribution. TABLE I CALCULATED I R DROPIN 0 . 1 M KCl (ID= limiting current; I, = reduction current; I, = charging current; Av = voltage correction) El/;vs. S.C.E. -0.420"

ID, pamp.

I,,

pamp.

I.,

ramp.

Ao,

mv.

17.0, mv.

16.0, mv.

0 113 107 113 78 58 58 59 38 37 40 0 15.5 21 16.7 141 12.7 14.0 15 85 11.3 15 11.2 120 200 4.9 5.2 0 4.5 200 2.8 2.6 2.6 200 44 0 4.2 26.7 121 3.0 8.7 200 ... 800 659 122 46 46 44 2000 993 118 63 65 64 ... 2000 778 0 49 46 49 2000 90 0 6.1 5.3 5.6 El/* values selected such that reduction occurs a t electrocapillary zero and charging current is negligible.

-0.850 -0.450" -1.250 -0.850 -1.100 -0.470" -0.470" -1.140

2000 2000 2000 2000 200

1816 915 636 220 175 139 83

... ...

Figures 2 and 3 show one of the solutions obtained by the process just described. In general,

0.5 1.0 Distance from orifice, cm. Fig. 3.-Calculated current density for A , the reduction current and B , the reduction and charging currents for a typical reduction situation : C, ideal, constant-voltage electrode (same conditions as Fig. 2.).

the effect of variable voltage is to flatten the current-density curve. The charging current is concentrated at the base of the stream and tends t o shift the reduction curve to the right. The calculation was carried out for several sets of conditions with results as summarized in Table I, Each line of the table represents a particular solution of the problem corresponding to a single point on a current-voltage curve. In each case the length of the stream from the orifice to the base of the tail-cone was taken as 1 cm. The first nine calculations assume reversible reduction, while the last four are for irreversible cases. The halfwave potential and limiting current corresponding to the particular assumed conditions are given in the first two columns; the reduction current, I,, and charging current, I,, obtained by summing the current densities, as finally adjusted, over the one cm. length, follow; and the corresponding voltage correction, that is, the difference between the measured cell voltage and the potential of the constantvoltage electrode that would draw the same reduction current appears in the fifth column.

.

5546

J. I